PROJECT SUMMARY/ABSTRACT Asymmetric or hemiparetic gait is a frequent and disabling consequence of stroke. Although hemiparetic gait is thought to stem from insufficient propulsive force generated by the paretic (i.e., weaker) leg, few interventions have targeted paretic leg propulsion for improvement. Moreover, existing gait training methods often require extensive assistance from trained personnel (i.e., physical therapists), expensive equipment (e.g., robotic exoskeletons, specialized treadmills, body-weight support), and the ability to travel to a rehabilitation center that offers these treatments. Given these barriers to accessing long-term gait rehabilitation, it is not surprising that gait impairments persist long-term following stroke and that most people are dissatisfied with their options for rehabilitation after leaving the hospital. The overall objective of this proposal is to evaluate a novel rehabilitation device called the Gait Propulsion Trainer (GPT) that offers targeted training to improve paretic leg propulsion during walking. It is low cost and easy to use, enabling future deployment in small clinical settings and in home- or community-based rehabilitation. The GPT is therefore be based on a scientific premise of gait impairment following stroke: that decreased propulsive force generation by the paretic leg significantly contributes to walking dysfunction. The GPT consists of a long cable attached to a harness worn around the hips; the other end of the cable is spooled on a reel attached to a stationary stand. The cable unspools behind the person as he or she walks over the ground. Triggered by force sensitive resistors taped on the shoe soles, resistance is applied to the unspooling cable from mid-to-terminal stance of the paretic leg. This system effectively resists forward movement of the participant?s body during paretic leg stance, requiring the person to push off with more force on the paretic side to keep moving forward. The aims proposed here will (1) test the ability of the GPT to increase propulsive force generation by the paretic leg, (2) identify relationships between GPT resistance and an individual?s response to training with the GPT, and (3) evaluate the effects of longer-term training with the GPT in a pilot study. The rationale for the current proposal is that, once we know the efficacy of the GPT and how to optimally deliver GPT therapy to individual people, we can then develop a personalized method for gait training that can be used by a large segment of the stroke population who want to improve their walking.